ResEncL-OpenMind-SimCLR Open-source Model - For Self-supervised Learning of 3D Medical Imaging Data

Resencl OpenMind SimCLR

Developed by AnonRes

The first comprehensive benchmark study model for self-supervised learning on 3D medical imaging data

3D Vision #3D Medical Imaging #Self-supervised Pretraining #Brain MRI

Downloads 16

Release Time : 5/6/2025

Model Overview

This model provides feature extraction models for 3D medical imaging pretrained with various self-supervised learning methods, primarily for brain MRI data analysis and processing

Model Features

Diverse Self-supervised Learning Methods

Provides 8 different self-supervised learning techniques for pretrained models, including VoCo, VF, MG, MAE, etc.

Dual Architecture Support

Offers both CNN-based ResEnc-L and Transformer-based Primus-M backbone architectures

Standardized Dataset

Pretrained on the OpenMind dataset, a large-scale, standardized public collection of brain MRI datasets

Medical Imaging Specialization

Specifically optimized for 3D medical imaging data, particularly suitable for brain MRI analysis

Model Capabilities

3D Medical Imaging Feature Extraction

Brain MRI Analysis

Medical Image Segmentation

Self-supervised Learning Pretraining

Use Cases

Medical Imaging Analysis

Brain MRI Segmentation

Used for segmentation tasks of brain MRI images

Can achieve good segmentation results through downstream fine-tuning

Medical Image Classification

Used for classification tasks of medical images

Pretrained models can serve as feature extractors

Medical Research

Self-supervised Learning Benchmarking

Used to compare the performance of different self-supervised learning methods on medical imaging

Provides pretrained models for 8 different methods

🚀 OpenMind Benchmark 3D SSL Models

This repository provides pre - trained checkpoints for 3D medical vision self - supervised learning from the OpenMind benchmark, facilitating downstream tasks in medical image analysis.

🚀 Quick Start

The models in this repository are pre - trained on the OpenMind Dataset. They are not recommended for direct feature extraction. Instead, use the downstream fine - tuning frameworks for segmentation and classification adaptation available in the adaptation repository.

⚠️ Important Note

While manual download is possible, we recommend using the auto - download feature of the fine - tuning repository by providing the repository URL on Hugging Face instead of a local checkpoint path.

📚 Documentation

Overview

This repository hosts pre - trained checkpoints from the OpenMind benchmark:
📄 An OpenMind for 3D medical vision self - supervised learning (Wald, T., Ulrich, C., Suprijadi, J., Ziegler, S., Nohel, M., Peretzke, R., ... & Maier - Hein, K. H. (2024).)
(arXiv:2412.17041) — the first extensive benchmark study for self - supervised learning (SSL) on 3D medical imaging data.

Each model was pre - trained using a particular SSL method on the OpenMind Dataset, a large - scale, standardized collection of public brain MRI datasets.

Model Variants

We release SSL checkpoints for two backbone architectures:

ResEnc - L: A CNN - based encoder [a, b]
Primus - M: A transformer - based encoder [Primus paper]

Each encoder has been pre - trained using one of the following SSL techniques:

Method	Description
Volume Contrastive (VoCo)	Contrastive pretraining method for 3D volumes
VolumeFusion (VF)	Spatial volume fusion - based segmentation SSL method
Models Genesis (MG)	Reconstruction and denoising based pretraining method
Masked Autoencoders (MAE)	Default reconstruction based pretraining method
Spark 3D (S3D)	Sparse reconstruction based pretraining mehtod (CNN only)
SimMIM	Simple masked reconstruction based pretraining method (TR only)
SwinUNETR SSL	Rotation, Contrastive and Reconstruction based pre - training method.
SimCLR	Transfer of 2D Contrastive learning baseline method to 3D

Additional Information

Model from the paper: An OpenMind for 3D medical vision self - supervised learning
Pre - training codebase used to create checkpoint: [MIC - DKFZ/nnssl](https://github.com/MIC - DKFZ/nnssl)
Dataset: AnonRes/OpenMind
Downstream (segmentation) fine - tuning: TaWald/nnUNet

OpenMind

📄 License

The model is licensed under CC - BY - 4.0.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご